Development of human ABO blood group A antigen on Escherichia coli Y1089 and Y1090

Expression of ABO or related antigenic carbohydrates on viral envelopes leads to neutralization in the presence of serum containing specific natural antibodies and complement

No definitive biologic function has been associated with the human ABO histo-blood group polymorphism, or any other terminal carbohydrate differences within or between closely related species. We have experimentally addressed the question of whether viral particles can become glycosylated as determined by the glycosylation (eg, ABO) status of the producer cell and as a result be affected by human serum containing specific natural antibodies (NAbs). Measles virus was produced in cells transfected with cDNA encoding, either human A-transferase, B-transferase, an inactive "O-transferase," or a pig alpha1-3galactosyltransferase (alpha1-3GT) synthesizing the Galalpha1-3Gal structure. The viruses were shown to carry the same ABO structures as the cells; that is, A but not B if produced in A-type cells, and B but not A if produced in B-type cells. Only O was detected on the virus produced from O-type cells, whereas reduced amounts of O appeared on the A- and B-type viral particles. In addition, the Galalpha1-3Gal structure was transferred onto measles only when grown in human cells expressing this structure. When subjected to human preimmune sera, the A-type, the B-type, and the Galalpha1-3Gal viral particles were partially neutralized in a complement-dependent manner. However, the O-type or the Galalpha1-3Gal-negative viral particles were not neutralized. The neutralization appeared to be mediated by specific NAb, as judged by specific inhibition using synthetic A and Galalpha1-3Gal oligosaccharides. Such viral glycosylation may thus partly explain why the ABO antigens and other similar intraspecies as well as interspecies polymorphic carbohydrates have evolved and been maintained over long evolutionary periods.

Blood group antigens: molecules seeking a function?

The blood group antigens have been dismissed by some researchers as merely 'icing on the cake' of glycoprotein structures. The fact that there are no lethal mutations and individuals have been described lacking ABO, H and Lewis antigens seems to lend weight to the argument. This paper reviews the research which suggests that these antigens do indeed have function and argues that blood group antigens play important roles in modulation of protein activity, infection and cancer. It explores the evidence and poses questions as to the relevance and implications of the results.